EVALUATION OF ESOPHAGEAL CYTOLOGY USING A NEURAL NET-BASED INTERACTIVE SCANNING SYSTEM (THE PAPNET SYSTEM) - ITS POSSIBLE ROLE IN SCREENINGFOR ESOPHAGEAL AND GASTRIC-CARCINOMA

A neural net-based, semiautomated, interactive computerized cell analy sis system (The PAPNET system, Neuromedical Systems, Suffern, NY) was used to examine cells from 138 esophageal smears obtained by lavage, b rushings, or balloon from as many patients. From each smear, trained h uman observers examined 128 cell images selected by the machine. Abnor mal cells were identified in all 35 patients with cancer, whether esop hageal, gastric, oral, or metastatic. Further, in 11 smears, the displ ayed images allowed the recognition of effects of radiotherapy and, in 14 smears, the diagnosis of a specific tumor type, such as squamous c ell carcinoma (8 patients) or adenocarcinoma (6 patients). In 3 additi onal cases, the diagnosis of ''carcinoma, not further specified,'' was established. One case of esophageal carcinoma in situ, not previously recognized on a smear or in the biopsy specimen, and one case of gast ric adenocarcinoma, not recognized in the smear, were identified in PA PNET-generated images. The possible application of the apparatus to th e triage of smears and population screening for esophageal and gastric carcinoma precursors is discussed.